Back

Understanding COVID-19 admissions in the UK; Analysis of Freedom of Information Requests

Jefferson, T.; Brassey, J.; Heneghan, C.

2022-06-06 infectious diseases
10.1101/2022.06.06.22276032 medRxiv
Show abstract

BackgroundThe progression and severity of the COVID 19 pandemic have been measured based on the daily and total numbers of cases, hospitalisations and deaths. We focused on the nature of hospitalisations from 2020 to 2022. MethodsWe analysed the role played by SARS-CoV-2 in the pandemic in the UK; we lodged FOI requests to Public Health Wales (PHW), Scotland (PHS), and Northern Ireland (PHA NI), the UK Health Security Agency (UKHSA) and NHS England. We asked for all-cause hospital admission monthly numbers reported by days of positivity to SARS-CoV-2 since admission from March 2020. We grouped replies by respondents. We considered any positive tests acquired from day 8 post admission as evidence of in hospital transmission. ResultsPHW, PHS and PHA NI, provided data within two months. The proportion of people admitted who became positive after eight or more days was 33% in Northern Ireland, 24% in Scotland and 45% in Wales. There are seasonal fluctuations reflecting community admissions but no evidence that the proportion of those infected in hospitals reduced over time. No authorities had viral load or symptoms data relating to their datasets. Given the limitations in PCR reporting, it is impossible to know how many "positive" cases were active. UKHSA did not hold the data, and NHS England did not clarify the content of its website. ConclusionAggregate data of "cases of Covid" in hospitals should not be used to inform policy of decision-makers until coordination, and proper interpretation of the dataset are instigated.

Matching journals

The top 3 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.